fix(routing): lower opt level for sliding_window.cu to avoid nvcc 13.0.3 ICE - #1452
Conversation
|
Auto-sync is disabled for draft pull requests in this repository. Workflows must be run manually. Contributors can view more details about this message here. |
|
Note Reviews pausedIt looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the Use the following commands to manage reviews:
Use the checkboxes below for quick actions:
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Enterprise Run ID: 📒 Files selected for processing (1)
📝 WalkthroughWalkthroughAdds a ChangesPer-file CUDA compile option for sliding_window.cu
Estimated code review effort🎯 1 (Trivial) | ⏱️ ~2 minutes 🚥 Pre-merge checks | ✅ 5✅ Passed checks (5 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing Touches🧪 Generate unit tests (beta)
Comment |
7c8e585 to
4dcbd6e
Compare
…_window.cu nvcc 13.0.3 on x86_64 crashes (signal 11) compiling sliding_window.cu across 7 GPU architectures — the recursive if constexpr template combined with the large multi-arch codegen exhausts compiler resources. --split-compile=0 breaks codegen into per-arch sub-jobs, sidestepping the crash with no impact on runtime behavior. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
4dcbd6e to
71e7334
Compare
|
/merge |
nvcc 13.0.3 on x86_64 segfaults (signal 11) at
-O3compilingsliding_window.cu— the ICE is triggered by the recursiveif constexprtemplate (permutation_cases). Dropping this file to-O2viaset_source_files_propertiessidesteps the bug with no runtime impact.